Inorg. Chem.
1995, 34, 4523—4524 4523
The Dianion of nt</o-Decaborane(14), nu/o-Dodecahydrodecaborate(2-), [BioHn2 ], and Its Solution
Behavior
Adam N. Bridges and Donald F. Gaines*
Department of Chemistry, University of Wisconsin—Madison, Madison, Wisconsin 53706
Received December 13, 1994
Our continuing interest in borane cage expansion via insertion
of boron and other main-group moieties has led us to investigate
the somewhat elusive nido- 2- anion, the most reactive
and least well characterized of the five known decaborane
anions: c/oso-BioHio2- (22 skeletal electrons, e~), nido-
, , 2- (24 e“), w'do-BioHi3~(24 e~), arachno-BioHh2- (26
e“), and arachno-BioHb~(26 e~). Our investigations of the
B10H122- anion and its chemistry using current high-field NMR
See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.
analyses led us to believe that the earlier report1 on the
characterization of unsolvated [PluAshBioHn and [PI13-
PMehBioHn may have contained characterization errors owing
to unsuspected side reactions (vide infra).
Downloaded via CINVESTAV on November 5, 2023 at 05:01:36 (UTC).
nido-B10H122- has proved to be difficult to characterize
structurally, as it is a very aggressive proton “sponge”, forming
nido-B10H13-, and will convert to other decaborane anions in
many coordinating solvents. These properties of nido- ß2-
do not appear to have been fully appreciated in prior studies.1·2
Herein we report improved syntheses of nido- ß2-. its high-
field NMR characterization, its decomposition under various
conditions, and some unexpected new reactions.
The original synthesis of NaatBioHn] was carried out in a
steel reactor over 6 days in diethyl ether.3 Yields of 75% were
reported, based on H2 evolution. Re-examinations of the
reaction, using both NaH and KH, have shown it to be very
dependent on both the hydride and the solvent employed.
Diglyme and Et20 are the most satisfactory,4 presumably
because the product is completely insoluble. In the preferred
diglyme solvent, the reaction is generally complete in 6 h at
room temperature and yields are quantitative, based on H2
evolution (eq l).5 The diethyl ether and diglyme insoluble
d‘slyme'·
B10H14 + excess MH
[M*0.7diglyme]2[B10H12] + 2H2 (1)
M =
Na, K
[Na-OJdiglymeMBioHu] is sparingly soluble in glyme (1,2-
(1) Youll, B.; Greenwood, N. N. J. Chem. Soc., Dalton Trans. 1975,158.
(2) Greenwood, N. N.; Sharrocks, D. N. J. Chem. Soc. A 1969, 2334.
(3) Wilkes, P. H.; Carter, J. C. J. Am. Chem. Soc. 1966, 88, 3441.
(4) Purity of the solvents is critical. Impurities in the solvents may not
only reduce yields but also alter the entire course of the reaction. This
is particularly true of diglyme (bp 162 °C), which is difficult to dry
and must always be distilled under vacuum.
(5) Under nitrogen in a 50 mL Schlenk vessel, equipped with a stir bar,
is dissolved 0.709 g (5.80 mmol) of decaborane(14) in about 25 mL
of dry diglyme. Then, 0.334 g (13.91 mmol) of oil-free NaH is added
slowly, cautiously in small amounts to the decaborane solution under
N2. The mixture effervesces copiously and becomes bright yellow.
After addition, the mixture is stirred for 12 h or is stirred until all the
yellow color has faded. The reaction mixture is filtered, and the solid
is washed several times with dry ether until it is gray-white. The
heterogeneous solid mixture (NaH and [Na0.7diglyme]2[BioHi2]) is
then extracted with THF to give a bright yellow filtrate. Immediate
vacuum evaporation of the THF extract leaves a free-flowing white
solid. The solid is collected using a fine frit and washed several times
with dry ether to give 1.919 g of [NaOJdiglymehtB 10H12] in a 93%
yield. For synthetic applications, the freshly prepared THF extract is
used directly.
0020-1669/95/1334-4523$09.00/0
dimethoxyethane, DME) and moderately soluble in THF and
CH3CN, forming yellow solutions, the color being attributed
to solvent complex formation.6 [KOJdiglymeMBioHn] dis-
solves only in CH3CN, also forming a yellow solution.7
The white, unsolvated sodium and potassium salts of the
BI0H,22- anion, prepared in Et20, and the corresponding
[MO.7diglyme]2[BioHi2] solvates, prepared in diglyme, appear
to be indefinitely stable in the absence of air and proton sources.
NMR confirms the presence of diglyme in the product, and
it appears to be associated with the metal ion. The diglyme
has no effect on the "B NMR spectrum or the reactivity of the
dianion in THF compared to the diglyme-free salts prepared in
Et20. The diglyme molecules remain associated with Na2BioHi2
in THF solution and are not removed by precipitation with Et20.
The diglyme can, however, be removed by vacuum pyrolysis,
during which decomposition occurs, giving products similar to
those observed by Wilkes and Carter.8
An intriguing aspect of MaiBioHn] solutions in THF, CH3-
CN, and DME is their yellow color, which we attribute to loose
solvent—BioHi22~ complex formation. Evaporation of these
solutions re-forms the original white, solvent-free M2IB10H12],
(or [M0.7diglyme]2[BioHi2]) salts. When isolation from THF
solution is followed by dissolution in CD3CN, the NMR
spectrum of the resulting yellow solution shows no evidence
of THF. (Unsolvated Na2[BioHi2] is precipitated from THF and
CH3CN solutions upon addition of Et20.)
The 160 MHz nB NMR spectrum of Na2BmHi2 in THF is
consistent with a high-symmetry solution structure (and very
unlike the spectrum reported1 for [PluAshBioHn). In CH3-
CN, the essentially identical spectrum is better resolved (Figure
1) and is readily assigned on the basis of nB—nB COSY NMR.
The spectrum is remarkably similar to that of BiqHb-, the most
noticeable difference being a single B(l,3) resonance in the
B,oH,22- spectrum. The considerable broadening of the B(6,9)
resonance, even in CH3CN, is tentatively attributed to solvent
interaction at those sites. Similar broadening is observed in
arachno- decaborane systems (i.e. BioHb*2L, BiqHb-‘L) in
which ligands are coordinated to the cage B(6,9) positions.
Strong cross-coupling observed between the B(6,9) and the
B(5,7,8,10) resonances in the nB—"B COSY NMR spectrum
(6) The amount of diglyme associated with the dianion was determined
by heating the solid in a vacuum to 200 °C, driving off the diglyme,
and weighing the diglyme retrieved. Although B10H122- decomposes
at this temperature, the products are nonvolatile borane anions and
H2, which do not affect the gravimetric measurement.
(7) [K0.7diglyme]2[BioH|2] can be prepared as above using slightly less
than 2 molar equiv of KH/mol of decaborane(14) in diglyme. In a
typical reaction, 0.965 g (7.90 mmol) of B10H14 is dissolved in 15
mL of dry diglyme and 0.606 g (15.15 mmol) of KH is added slowly
to the solution under N2. After much effervescence, the yellow solution
is allowed to stir briskly for 4 days, during which time a white
precipitate forms. The mixture is then filtered through a fine frit and
the solid is washed with diglyme, Et2Ü, and then several portions of
dry THF to leave 1.888 g of a white, hydride-free product mixture of
[K0.7diglyme]2[BioHi2] ,74% along with 13% of both BioHm2- and
BioHi42~ salts (nB NMR analysis of the totally soluble reaction
mixture in CH3CN). The actual yield of [K-0.7diglyme]2[BioHi2] is
46%.
(8) Carter, J. C; Wilkes, P. H. Inorg. Chem. 1970, 9, 1777.
&copy; 1995 American Chemical Society
4524 Inorganic Chemistry, Vol. 34, No. 18, 1995
1C.0 0.0 -10.0 -20.0
Figure 1. The 160 MHz nB NMR spectrum of Na2[BmHi2] in CH3-
CN showing resonance assignments with positive downfield shifts
referenced to BF3-Et20.
of BioHi22~ augurs well for the absence of bridging hydrogens
spanning these positions [B(5—6), B(6—7), B(8—9), B(9—10)].
In the 499 MHz spectrum of Na2[BioHi2] in CD3CN
solution,9 the unusually high field shift for the bridging
hydrogens of intensity 2 at (ó) = —4.2 ppm is similar to that
observed for the bridging hydrogens spanning the B(7—8) and
B(5—10) vertices in arachno-B\qH 12-2L, and arachno-BioH|42~.
These data and the instability (vide infra) of the yellow
M2[BioH|2] solutions in CH3CN, THF, and DME lead us to
suggest that they contain loose solvent complexes of the
B,0H,22- anion as shown in the following scheme:
NaH L
Na2[B10H12] ^
Et2G
I transfer mechanism remains unknown. Another unexpected
b10h14 L =
THF, DME, MeCN
NaH
I
[Na«0.7 DG]2Bi0Hi2
DG
We further suggest that these are most likely BioH|2*2L2_
complexes having an overall structure similar to that of arachno-
BioH]2-2L and arachno-B]()Hif~, while having a skeletal
electron count (28 e") of a hypho dianion cluster fragment, as
shown below. The instability of this complex is similar to that
observed for the only fully characterized hypho dianion,
B5H112-:10
Aside from this arrangement, which is the only static possibility
that preserves the observed symmetry, the hydrogens could be
in flux around the cage face, as suggested by Todd11 to explain
the high symmetry of the " NMR spectra of 10 13- solutions,
but such a process is not consistent with the nB NMR and 'B—
"B COSY NMR analysis (vide supra). Another plausible
explanation is that a very labile arachno-B qH 12-L2~ solution
species exists in which monobase dissociation/association occurs
rapidly on the NMR time scale at the B(6,9) positions. Such a
(9) {" } NMR at 499 MHz in CD3CN (B-H region only): ó = +2.4
(2Ht), +1.9 (6Ht), -0.6 (2Ht), -4.2 (2H„). The spectrum is available
as supporting information.
(10) McGaff, R. W.; Gaines, D. F. Inorg. Chem. 1995, 34, 1009.
(11) Todd, L. J.: Bodner, G. M.; Siedle. A, R. J. Inorg. Nucí. Chem. 1971,
33, 3671.
Communications
species avoids consideration of an unprecedented hypho-
decaborane dianion12 but could still provide the driving force
for the disproportionation and decomposition of BmHi22~
solutions (vide infra).
The solution species, [BioHi22~]tzL (L = DME, THF, CH3-
CN), are not stable. In DME, the sparingly soluble Na2BioH]2
converts entirely to BioHio2- over a 24 h period. In contrast,
THF solutions of Na2[BioHi2] slowly convert to a 1:1 mixture
-30.0 -40.0 of c/oso-BioHio2- and arachno-B10H142-, as well as closo-
B9H92-. These species have also been observed as decomposi-
tion products in the vacuum pyrolysis8 of Na2BioHi2. The
CH3CN solutions of Na2[BioH]2] and K2[Bi0Hi2] decompose
to a more complex mixture of products, the most prominent of
which are B9H92- and BioHio2-.
Attempts to isolate crystalline salts of BioH|22_ via metathesis
with various large, cationic salts (i.e. [Pli3PMe]Br, [PtuAsjCl,
[Bu4N]Br) in a THF/CH2C12 solution did not produce the
expected BioH)22_ species as reported by Greenwood,1 but rather
a mixture of products, including a 1:1 mixture of B10H102"" and
BioH]42~. While the yield of these dianions varied with the
particular cation employed, the 1:1 ratio was always observed,
and the introduction of CH2C12 solutions of the phosphonium
and arsonium salts accelerated this conversion process. The
observation of simultaneous formation of B10H142- and BuHio2'
leads irresistibly to the suggestion that two hydrogen atoms
w
B10Hi2*nL‘2 transfer from one BmH|22~ group to another, though the actual
product isolated from these reaction mixtures is the monocarbon
carborane anion BioHnCH-. Indeed, reactions with Na2Bio-
--
B10H12*nL4 H,2-2THF and CH2X2 (X = Br, I) in THF produced a 1:1
mixture of BioHi2CH~ and B10H13- exclusively,13 which leads
us to suggest that BioH]22_ is acting both as a receptacle for
CH2X2 insertion and as a proton acceptor. This carbon insertion
route to the BioHi2CH~ anion may prove to be more convenient
than the multistep synthesis currently available for BioHi2CH-.14
Acknowledgment. We thank the National Science Founda-
tion for support of this research and for major departmental
instrumentation grants.
Supporting Information Available: Figures of the "B-"B COSY
NMR and 'HC'Bj NMR spectra of [Na-0.7diglyme]2[BloH12] in CD3-
CN solution (3 pages). Ordering information is given on any current
masthead page.
IC941421I
(12) We thank a reviewer for suggesting this alternate explanation.
(13) In a typical reaction, 1.64 mmol of [NaO.7diglyme]2[BjoH|2] is
prepared from 0.20 g( 1.6 mmol) of B10H14 and 0.15 g(6.2 mmol) of
NaH as described above and dissolved in 35 mL of THF under N2.
Then 1.20 g (6.90 mmol) of CH2Br2 is added to the solution. The
reaction mixture is stirred at room temperature overnight, during which
time a white precipitate forms. Analysis by nB NMR shows two
products: B10H13- and BioHi2CH~. NaH is then added to the reaction
mixture and stirring is continued for an additional 12 h. NaH is added
in order to regenerate the BioH]22_ anion for further insertion of CH2-
Br2. Subsequent analysis by "B NMR shows BioHnCH- as the major
1
product with smaller amounts of B10H13- and some other impurities.
The solution is filtered, and the filtrate is acidified with 1.3 mL of 1
M HCl/Et20. The solvent and excess acid are immediately removed
1
under vacuum. The residue is kept under vacuum overnight in order
to remove most of the B10H14. The residue is then extracted with
several portions of hot hexanes and redissolved in THF. Then 0.143
g (1.31 mmol) of Me4NCl is added to the solution, and the mixture is
stirred for several hours. The solution is filtered again to remove NaCl,
and most of the THF/Et2G is removed to yield a bright-yellow liquor.
The product is precipitated with Et20, filtered off, and recrystallized
from THF/Et2G to yield 0.089 g (0.403 mmol) of Me4N[BioHi2CH]
in a 25% yield. nB NMR shows trace contamination of B9H14-.
(14) Knoth, W. H.; Little, J. L.; Lawrence. J. R.; Scholer, F. R.; Todd. L.
J. Inorg. Synth. 1968, II, 33.